@olib-ai/owl-browser-mcp 1.0.2 → 1.0.4

package/dist/index.js

@@ -22529,6 +22529,204 @@ var openapi_default = {
         }
       }
     },
+    "/api/execute/browser_extract_site": {
+      post: {
+        summary: "Browser Extract Site",
+        description: "Extract content from multiple pages of a website. Crawls links starting from a URL and extracts content in the specified format. Returns a job_id immediately for async progress tracking. Use browser_extract_site_progress to monitor and browser_extract_site_result to get output.",
+        tags: [
+          "General"
+        ],
+        requestBody: {
+          required: true,
+          content: {
+            "application/json": {
+              schema: {
+                type: "object",
+                properties: {
+                  context_id: {
+                    type: "string",
+                    description: "The unique identifier of the browser context (e.g., 'ctx_000001')"
+                  },
+                  url: {
+                    type: "string",
+                    description: "Starting URL to begin extraction from"
+                  },
+                  depth: {
+                    type: "integer",
+                    description: "How many link levels to follow from the starting page. Default: 2. Higher values extract more pages but take longer"
+                  },
+                  max_pages: {
+                    type: "integer",
+                    description: "Maximum number of pages to extract. Default: 5. Limits total extraction to prevent runaway crawling"
+                  },
+                  follow_external: {
+                    type: "boolean",
+                    description: "Whether to follow links to external domains. Default: false. When false, only links within the same domain are followed"
+                  },
+                  output_format: {
+                    type: "string",
+                    description: "Output format for extracted content: 'markdown' (default), 'text', or 'json'. Markdown preserves structure, text is plain, JSON includes metadata",
+                    enum: [
+                      "markdown",
+                      "text",
+                      "json"
+                    ]
+                  },
+                  include_images: {
+                    type: "boolean",
+                    description: "Include resolved image URLs in output. Default: true"
+                  },
+                  include_metadata: {
+                    type: "boolean",
+                    description: "Include page title and description metadata. Default: true"
+                  },
+                  exclude_patterns: {
+                    type: "string",
+                    description: `JSON array of URL patterns to skip (glob patterns). Example: '["*/login*", "*/admin/*"]'`
+                  },
+                  timeout_per_page: {
+                    type: "integer",
+                    description: "Timeout per page in milliseconds. Default: 10000 (10 seconds)"
+                  }
+                },
+                required: [
+                  "context_id",
+                  "url"
+                ]
+              }
+            }
+          }
+        },
+        responses: {
+          "200": {
+            description: "Successful response"
+          },
+          "400": {
+            description: "Bad request"
+          },
+          "401": {
+            description: "Unauthorized"
+          }
+        }
+      }
+    },
+    "/api/execute/browser_extract_site_progress": {
+      post: {
+        summary: "Browser Extract Site Progress",
+        description: "Get progress of a site extraction job. Returns pages_completed, pages_total, current_url, and status. Status can be: 'running', 'completed', 'cancelled', or 'error'.",
+        tags: [
+          "General"
+        ],
+        requestBody: {
+          required: true,
+          content: {
+            "application/json": {
+              schema: {
+                type: "object",
+                properties: {
+                  job_id: {
+                    type: "string",
+                    description: "The job ID returned from browser_extract_site"
+                  }
+                },
+                required: [
+                  "job_id"
+                ]
+              }
+            }
+          }
+        },
+        responses: {
+          "200": {
+            description: "Successful response"
+          },
+          "400": {
+            description: "Bad request"
+          },
+          "401": {
+            description: "Unauthorized"
+          }
+        }
+      }
+    },
+    "/api/execute/browser_extract_site_result": {
+      post: {
+        summary: "Browser Extract Site Result",
+        description: "Get the result of a completed site extraction job. Returns the formatted content based on the output_format specified when starting the job (markdown, text, or json).",
+        tags: [
+          "General"
+        ],
+        requestBody: {
+          required: true,
+          content: {
+            "application/json": {
+              schema: {
+                type: "object",
+                properties: {
+                  job_id: {
+                    type: "string",
+                    description: "The job ID returned from browser_extract_site"
+                  }
+                },
+                required: [
+                  "job_id"
+                ]
+              }
+            }
+          }
+        },
+        responses: {
+          "200": {
+            description: "Successful response"
+          },
+          "400": {
+            description: "Bad request"
+          },
+          "401": {
+            description: "Unauthorized"
+          }
+        }
+      }
+    },
+    "/api/execute/browser_extract_site_cancel": {
+      post: {
+        summary: "Browser Extract Site Cancel",
+        description: "Cancel a running site extraction job. Returns success status.",
+        tags: [
+          "General"
+        ],
+        requestBody: {
+          required: true,
+          content: {
+            "application/json": {
+              schema: {
+                type: "object",
+                properties: {
+                  job_id: {
+                    type: "string",
+                    description: "The job ID to cancel"
+                  }
+                },
+                required: [
+                  "job_id"
+                ]
+              }
+            }
+          }
+        },
+        responses: {
+          "200": {
+            description: "Successful response"
+          },
+          "400": {
+            description: "Bad request"
+          },
+          "401": {
+            description: "Unauthorized"
+          }
+        }
+      }
+    },
     "/api/execute/browser_extract_json": {
       post: {
         summary: "Browser Extract Json",
@@ -26797,6 +26995,7 @@ var openapi_default = {
 var API_ENDPOINT = process.env.OWL_API_ENDPOINT || "http://127.0.0.1:8080";
 var API_TOKEN = process.env.OWL_API_TOKEN || "";
 var activeContexts = /* @__PURE__ */ new Map();
+var toolIntegerFields = /* @__PURE__ */ new Map();
 function convertOpenAPIToMCPTools(schema) {
   const tools = [];
   for (const [path, pathItem] of Object.entries(schema.paths)) {
@@ -26810,6 +27009,7 @@ function convertOpenAPIToMCPTools(schema) {
     };
     if (requestSchema?.properties) {
       const properties = {};
+      const integerFields = /* @__PURE__ */ new Set();
       for (const [propName, propDef] of Object.entries(requestSchema.properties)) {
         const prop = {};
         if (propDef.type === "string") {
@@ -26821,6 +27021,9 @@ function convertOpenAPIToMCPTools(schema) {
           prop.type = "boolean";
         } else if (propDef.type === "number" || propDef.type === "integer") {
           prop.type = "number";
+          if (propDef.type === "integer") {
+            integerFields.add(propName);
+          }
         } else if (propDef.type === "array") {
           prop.type = "array";
           if (propDef.items) {
@@ -26838,6 +27041,9 @@ function convertOpenAPIToMCPTools(schema) {
         properties[propName] = prop;
       }
       inputSchema.properties = properties;
+      if (integerFields.size > 0) {
+        toolIntegerFields.set(toolName, integerFields);
+      }
     }
     if (requestSchema?.required && requestSchema.required.length > 0) {
       inputSchema.required = requestSchema.required;
@@ -26850,6 +27056,19 @@ function convertOpenAPIToMCPTools(schema) {
   }
   return tools;
 }
+function coerceIntegerFields(toolName, args) {
+  const integerFields = toolIntegerFields.get(toolName);
+  if (!integerFields || integerFields.size === 0) {
+    return args;
+  }
+  const coerced = { ...args };
+  for (const field of integerFields) {
+    if (field in coerced && typeof coerced[field] === "number") {
+      coerced[field] = Math.round(coerced[field]);
+    }
+  }
+  return coerced;
+}
 async function callBrowserAPI(toolName, args) {
   const url2 = `${API_ENDPOINT}/api/execute/${toolName}`;
   const headers = {
@@ -26858,11 +27077,12 @@ async function callBrowserAPI(toolName, args) {
   if (API_TOKEN) {
     headers["Authorization"] = `Bearer ${API_TOKEN}`;
   }
+  const coercedArgs = coerceIntegerFields(toolName, args);
   try {
     const response = await fetch(url2, {
       method: "POST",
       headers,
-      body: JSON.stringify(args)
+      body: JSON.stringify(coercedArgs)
     });
     const contentType = response.headers.get("content-type") || "";
     let data;
@@ -26890,8 +27110,18 @@ async function callBrowserAPI(toolName, args) {
     };
   }
 }
+function isContextLimitError(result) {
+  if (typeof result === "object" && result !== null) {
+    const obj = result;
+    return obj.error === true && obj.code === "CONTEXT_LIMIT_EXCEEDED";
+  }
+  return false;
+}
 function trackContext(toolName, args, result) {
   if (toolName === "browser_create_context" && typeof result === "object" && result !== null) {
+    if (isContextLimitError(result)) {
+      return;
+    }
     const contextId = result.context_id;
     if (contextId) {
       activeContexts.set(contextId, { createdAt: /* @__PURE__ */ new Date() });
@@ -26921,6 +27151,33 @@ function formatResponse(toolName, result) {
     };
   }
   const data = result.data;
+  if (typeof data === "object" && data !== null) {
+    const dataObj = data;
+    if (dataObj.success === true && typeof dataObj.result === "object" && dataObj.result !== null) {
+      const innerResult = dataObj.result;
+      if (innerResult.error === true && innerResult.code === "CONTEXT_LIMIT_EXCEEDED") {
+        const details = innerResult.details || {};
+        const errorMsg = innerResult.message || `Developer license context limit reached (${details.max_contexts || "unknown"} contexts). Close existing contexts or upgrade license.`;
+        return {
+          content: [
+            {
+              type: "text",
+              text: JSON.stringify({
+                success: false,
+                error: errorMsg,
+                code: "CONTEXT_LIMIT_EXCEEDED",
+                details: {
+                  current_contexts: details.current_contexts,
+                  max_contexts: details.max_contexts,
+                  license_type: details.license_type || "developer"
+                }
+              }, null, 2)
+            }
+          ]
+        };
+      }
+    }
+  }
   const isImageTool = toolName === "browser_screenshot" || toolName === "browser_get_live_frame";
   if (isImageTool) {
     let base64Data = null;

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@olib-ai/owl-browser-mcp",
-  "version": "1.0.2",
+  "version": "1.0.4",
   "description": "MCP server for Owl Browser HTTP API - 144 browser automation tools with anti-detection",
   "type": "module",
   "main": "dist/index.js",